Abstract: DESCRIPTION: (adapted from the application): The project focuses on the development, and applications, of models and estimation procedures to analyze human survival at late ages in longitudinal and survival data for biologically related and unrelated individuals as well as survival data obtained in stress experiments with laboratory insects. The central idea which unifies the approaches is associated with the concept of hidden frailty. This idea will be further developed to analyse the effects of genetic and environmental influence on mortality and longevity using models of correlated fixed frailty, quadratic hazard models, models of changing frailty, and models of lpartly observed correlated health histories. Frailty models will also be developed to separate the effects of hormesis and heterogeneity in survival data. The methods will be applied to survival data from the Danish Twin Registry (DTR), the Longitudinal Studies of Aging in Danish Twins (LSADT) (collected in project 2), the 1982-1994 National Long Term Care Surveys (NLTCS; available at DU) and to data from stress-experiments on laboratory insects done in project 4. Five specific aims will be addressed: (1) Estimate the genetic contribution to individual susceptibility to disease and death using proportional hazard frailty models applied to DTR data; (2) Estimate the genetic contribution to individual susceptability to disease and death using the bivariate quadratic hazard model applied to DTR dam; (3) Evaluate current trends in disability and chronic conditions by estimating parameters of individual health histories in selected cohorts of NLTCS and LSADT data; (4) Evaluate the/nfiuence of genetic and environmental factors on disability, morbidity, and mortaity in DTR and LSADT data; (5) Evaluate the effects of hormesis and heterogeneity in the analysis of stress-experiment data on populations of Drosoinvestigatorzila melanogaster.